A team of researchers has engineered a common tobacco relative to produce five distinct psychedelic compounds simultaneously, pulling genetic instructions from fungi, plants, and animals and stitching them into a single organism. The work, carried out in the lab plant Nicotiana benthamiana, yielded psilocin, psilocybin, DMT, bufotenin, and 5-MeO-DMT in the same leaves. The result is the most chemically diverse plant-based psychedelic production system reported to date, and it arrives as clinical interest in these substances for treating depression and PTSD continues to grow.
Nine Genes, Three Kingdoms, One Plant
The central achievement is a feat of cross-kingdom genetic assembly. With the addition of nine genes sourced from mushrooms, various plant species, and animal pathways, the engineered tobacco plants produced psilocin and psilocybin, compounds normally found in psychoactive mushrooms, alongside DMT from various plants, bufotenin, and 5-MeO-DMT. The peer-reviewed study, titled “Complete Biosynthesis of Psychedelic Tryptamines from Three Kingdoms in Plants,” was described in a Science Advances paper that details the pathway reconstruction logic and the specific enzyme set used to coax the tobacco plant into manufacturing compounds it would never produce on its own.
What makes this different from earlier single-compound experiments is the breadth. Previous efforts in synthetic biology typically focused on one psychedelic at a time, often in microbial hosts like yeast or E. coli. Producing five tryptamine-class psychedelics in a single plant chassis required the researchers to coordinate enzymes that evolved in entirely separate branches of life, a design challenge that demanded careful sequencing of biochemical steps so that intermediates flowed correctly from one reaction to the next. Each added gene had to be expressed at the right level, in the right cellular compartment, and in a compatible order, or the pathway would stall or shunt metabolites into dead ends.
To manage that complexity, the team divided the overall pathway into modular segments corresponding to different biosynthetic tasks: tryptamine formation, hydroxylation, phosphorylation, and methylation. By testing combinations of these modules in transiently transformed leaves, they could iteratively refine the construct until all five target molecules appeared in measurable quantities. The resulting plants do not resemble “psychedelic crops” in any visible way, but chemically, their leaves contain a cocktail of psychoactive tryptamines that no natural organism is known to produce together.
AlphaFold3 and the Enzyme Puzzle
Part of what enabled the work was computational protein prediction. The researchers made use of AlphaFold3-based modeling, an artificial intelligence approach that predicts the three-dimensional structure of a protein from its amino acid sequence. That capability matters because enzymes from distantly related organisms do not always fold or function predictably when expressed in a new host. Structural predictions helped the team identify which candidate enzymes were most likely to remain active inside tobacco leaf cells, reducing trial-and-error cycles in the lab.
In particular, enzymes responsible for late-stage modifications, such as methyltransferases and phosphotransferases, can be finicky, losing activity when removed from their native cellular context. By comparing predicted structures and active sites, the researchers could prioritize variants that were more likely to tolerate the plant’s intracellular environment. This rational selection did not eliminate the need for empirical testing, but it narrowed the search space dramatically and increased the odds that each introduced gene would contribute to the final product profile.
The approach also extended to engineering halogenated analogs of the target psychedelics, according to the project’s BioProject entry in a public sequence database. Halogenation (the addition of chlorine or bromine atoms to a molecule) can alter a drug’s pharmacology, potentially improving receptor selectivity or metabolic stability. Demonstrating that the same plant platform can generate both natural and modified versions of these compounds hints at a flexible production system rather than a one-off proof of concept. In principle, the same design rules could be used to build libraries of psychedelic analogs for preclinical screening.
Building on the Mescaline Pathway
The new study did not emerge from a vacuum. The same Weizmann Institute research group, led by Berman and Aharoni, had previously demonstrated the reconstruction of the mescaline pathway in Nicotiana benthamiana and in yeast, as reported in a plant biology journal. That earlier work established the lab’s methodology for transplanting complex alkaloid pathways into organisms that do not naturally produce them, including strategies for balancing flux through multiple enzymatic steps and preventing the accumulation of toxic intermediates.
Mescaline, a phenethylamine alkaloid from cacti, presents a different biosynthetic architecture than tryptamine-based psychedelics such as psilocybin and DMT. Moving from a single phenethylamine to five tryptamines therefore represents a significant expansion of the platform’s chemical reach. It also validates the group’s iterative strategy: by reusing the same plant host and transformation techniques while swapping in new sets of pathway genes, they can progressively tackle more ambitious molecular targets without reinventing their experimental toolkit each time.
This continuity matters for scalability. Once a host species and delivery method are well characterized, later projects can focus on pathway design rather than basic troubleshooting. The mescaline work provided a template for promoter selection, subcellular targeting signals, and cofactor balancing, all of which could be adapted to the new tryptamine pathways. The five-psychedelic plant is thus both a scientific milestone and a proof that the platform can be extended to other psychoactive or therapeutic small molecules.
Why Plants Instead of Microbes?
Most coverage of synthetic psychedelic production has focused on yeast and bacteria, which grow fast and are easier to contain in bioreactors. So why bother with plants? The researchers argue that using plants as production hosts could be simpler and more sustainable than existing processes, making it easier to research therapeutic uses. Plants supply their own carbon fixation, water uptake, and cellular compartments for storing toxic intermediates, which means they do not require expensive fermentation media or sterile steel tanks.
Nicotiana benthamiana, in particular, grows quickly, accepts foreign genes readily through a well-established infiltration technique, and has been used for decades as a workhorse in plant molecular biology. Its leaves can be transiently transformed, producing target proteins or metabolites within days rather than the weeks required to build and optimize stable transgenic lines. That speed allows researchers to test multiple pathway designs in parallel, an advantage when tuning a complex multi-step biosynthetic network.
A recent review in a biotechnology journal surveyed the broader field of psychedelic pathway engineering, cataloging microbial and plant-based approaches for DMT, 5-MeO-DMT, bufotenin, and psilocybin production. The review’s conclusion aligns with the Weizmann team’s bet: for complex, multi-enzyme pathways that involve components from different kingdoms of life, plant hosts can outperform microbial alternatives because their cellular machinery is better equipped to fold and modify eukaryotic proteins and to compartmentalize reactive intermediates.
That does not mean plants will replace fermenters for every application. Microbes still offer advantages in tightly controlled industrial settings and for simpler pathways. But for compounds that require dozens of coordinated enzymatic steps or that benefit from sequestration in vacuoles and other plant-specific organelles, leafy bioreactors may prove more tractable. The five-psychedelic tobacco plants showcase exactly that niche: a dense, modular pathway assembled from fungi, plants, and animals, running smoothly inside a familiar greenhouse species.
From Lab Curiosity to Therapeutic Toolkit
The immediate impact of this work is scientific rather than clinical. The quantities of each psychedelic produced in the experimental plants remain modest, and regulatory barriers around psychoactive compounds would complicate any near-term commercial deployment. Still, the demonstration that a single plant can be programmed to synthesize multiple psychedelics, and even modified analogs, opens new avenues for drug discovery and comparative pharmacology.
Researchers could, for example, generate panels of related tryptamines in the same host and directly compare their receptor binding profiles or behavioral effects in animal models. Because the underlying genetic constructs are modular, swapping in alternative tailoring enzymes or introducing subtle mutations could yield dozens of candidate molecules with distinct properties. For conditions like depression, anxiety disorders, or PTSD, where psychedelic-assisted therapy is already under investigation, such a library could help disentangle which structural features correlate with therapeutic benefit versus undesirable side effects.
There are also broader implications for how society sources psychoactive and therapeutic compounds. Traditional extraction from wild or cultivated organisms can be slow, variable, and ecologically disruptive, while fully synthetic routes may involve hazardous reagents and generate significant waste. Engineered plants offer a third path: living factories that grow under sunlight, water, and basic nutrients, yet are customized at the genetic level to produce specific, high-value molecules.
For now, the five-psychedelic tobacco plants remain confined to controlled research facilities, where they serve as a striking illustration of what modern synthetic biology can do. As techniques like AI-guided enzyme design, modular pathway assembly, and plant-based expression systems continue to mature, similar strategies could be applied to everything from painkillers and anticancer agents to flavor compounds and fragrances. The convergence of these tools suggests a future in which the chemical diversity of nature is not just cataloged, but actively reassembled, one engineered leaf at a time.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
